INTRODUCTION: The biological mechanisms through which increased physical activity or structured exercise training lowers the risk of recurrent cardiac events are incompletely understood. We examined the extent to which modification of primary risk markers explains the association between physical activity and cardiovascular death in participants with diagnosed cardiovascular disease (CVD). METHODS AND RESULTS: In a prospective study of 1429 participants with physician-diagnosed CVD living in England and Scotland (age = 66.5 ± 11.1 yr (mean ± SD), 54.2% men), we measured physical activity and several risk markers (body mass index, total-to-HDL cholesterol ratio, diagnosed diabetes, systolic blood pressure, resting heart rate, C-reactive protein) at baseline. The main outcome was CVD death. There were a total of 446 all-cause deaths during an average of 7.0 ± 3.1 yr of follow-up, of which 213 were attributed to cardiovascular causes. Participation in moderate to vigorous physical activity at least three sessions per week was associated with lower risk of CVD death (hazard ratio = 0.61, 95% confidence interval = 0.38-0.98). Physically active participants demonstrated significantly lower levels of body mass index, diabetes, and inflammatory risk (C-reactive protein). Metabolic (body mass index, total-to-HDL cholesterol ratio, and physician-diagnosed diabetes) and inflammatory risk factors explained an estimated 12.8% and 15.4%, respectively, of the association between physical activity and CVD death. CONCLUSIONS: Physical activity may reduce the risk of secondary CVD events, in part, by improving metabolic and inflammatory risk markers.